Study period and setting
The study was conducted in selected public hospitals in Addis Ababa, the capital of Ethiopia, from May 1, 2023, to March 30, 2024. Among the twelve government hospitals, three hospitals, namely, Zewditu Memorial Hospital, St. Paul’s Millennium Hospital Medical College and Menelik Hospital, were selected via a simple random sampling technique. Pregnant women attending ANC units in the aforementioned hospitals were the study participants. The city has a projected population of 3,603,000, and 49.97% are females, 34.4% of whom are in the reproductive age group, according to the 2019 CSA projection [8]. The city lies at an elevation of 2355 meters above sea level.
Study Design
An institution-based prospective follow-up study design was used for pregnant women attending antenatal care units in the selected public hospitals in Addis Ababa from May 1, 2023, to March 30, 2024.
Population
Source population
All pregnant women above 18 years of age who have ANC booked at the selected public hospitals in Addis Ababa, Ethiopia, were included.
Study population
All randomly selected pregnant women who were above 18 years of age had ANC booking during the time of data collection at each selected public hospital in Addis Ababa, Ethiopia.
Eligibility criteria
Inclusion criteria
Women with singleton pregnancies above 18 years of age signed the written informed consent form, started IFAS with Hb levels > 7 mg/dl, and had ANC follow-up at the selected public hospitals during the time of data collection in Addis Ababa, Ethiopia.
Exclusion criteria
Women with severe anemia initially and throughout pregnancy (Hb < 7 mg/dl), blood transfusion, smoking, comorbid conditions, including but not limited to HTN, DM, CKD, known hematologic disorders, and infections, including helminthiasis, HIV and malaria, were included.
Sampling technique and procedure
On the basis of previous hospital records, the total number of pregnant women who had ANC follow-up at each selected hospital (within a month before the start of data collection) was obtained. This was followed by a proportional allocation of the number of ANC attendees to the respective hospitals. Using the ANC registry in the respective hospitals as a sampling frame, a systematic random sampling method was employed to select pregnant women who had ANC booking at the selected public hospital (Fig. 1). Pregnant women receiving ANC services at each selected public hospital during the data collection/study period who met the inclusion criteria had a random chance of being approached. The interviews were performed until the required sample size of 410 was achieved. Consent was secured verbally and in written form before the start of the data collection.
Variables of the study
Dependent variable
Maternal blood hemoglobin levels.
Independent variable
Sociodemographic factors (age, occupation, educational status, marital status, household size), MUAC, birth interval, parity, stage of pregnancy, maternal blood group, timing of ANC booking, IFAS adherence, duration of intake of IFAS, planned or unplanned pregnancy, and dietary diversity score.
Operational definitions
Anemia was defined as a hemoglobin value of less than 11 g/dL [9]. Hemoglobin responses to IFA supplementation were categorized as either adequate or inadequate [5, 10]. The criteria for these categories were as follows:
Adequate response: An increase in the hemoglobin value of at least 1 g/dL after a minimum of one month of supplementation, indicative of iron deficiency [11, 12].
Inadequate response: A change in the hemoglobin value of less than 1 g/dL after a minimum of one month of IFA supplementation. This response suggests the presence of functional iron deficiencies [5].
The normal reference range for hematological parameters, depicted below, was used: [13]. The severity of anemia was classified as follows: mild anemia (Hb: 10–10.9 g/dl), moderate anemia (Hb: 7–9.9 g/dl), and severe anemia (Hb < 7.0 g/dl) [1]. Gestational ages were categorized into the first trimester (1–14 weeks), second trimester (15–28 weeks) and third trimester (29 and above) using WHO classification criteria. Pregnant women with a MUAC value of 23–33 cm were considered to have normal nutritional status; those with MUAC measurements < 23 cm were categorized as undernourished, and those above 33 cm were categorized as obese [14, 15].
Compliance with the IFA supplement was assessed on the basis of pill count and self-reported methods. Women who took 70% or more of the IFA tablets, equivalent to taking at least 5 days a week throughout the study period, were considered adherent [16], using recording, self-reporting, pill counting and checking their cards. Otherwise, participants were considered nonadherent and excluded from the study.
Dietary intake was assessed via a food frequency questionnaire using the minimum dietary diversity for women (MDD-W), which was adopted by the Food and Agricultural Organization of the United Nations. The minimum dietary diversity for women (MDD-W) is a population-level indicator of diet diversity validated for women aged 15–49 years. It is a dichotomous indicator based on 10 food groups consumed locally (in this case, in Ethiopia) and is considered the standard for measuring population-level dietary diversity in women of reproductive age [17]. Pregnant women who consumed five or more food items in the last 24 hours, out of the 10 food groups, were considered to have adequate dietary diversity; otherwise, they were considered to have poor dietary diversity (Fig. 2).
Data collection procedures, tools and quality control
Data collection tools
A structured questionnaire was prepared in English and Amharic. The Amharic version was completed by the participants if they were able to read and write in the Amharic language; otherwise, the data collector completed the questionnaire by asking the respondents. The response was subsequently translated back into English. The questions explored and addressed the respondent’s sociodemographic profiles, data on parity/gravidity, intake of IFAS and dietary diversity.
Data collection procedure and quality control
Pregnant women who met the inclusion criteria were prescribed 60 mg of elemental iron plus 400 µg of folic acid oral tablets once daily if nonanemic and 120 mg of elemental iron plus 800 µg of folic acid once daily if anemic. The blood hemoglobin level of the participants was measured two times, i.e., at the first ANC visit (baseline, before the start of IFA supplementation) and after the intake of iron folic acid tablets, at the endpoint of the follow-up. All the respondents were taking similar iron‒folic acid tablets throughout the follow-up period.
A 5 mL venous blood sample was collected from respondents while they were seated comfortably, using iron-free heparinized test tubes. Professional nurses conducted the procedures at each respective hospital. Hematological assessments were performed at both the baseline and the end point (after at least 4 weeks of IFAS supplementation). All standard precautions were strictly adhered to during and after blood collection. Hematological analysis was conducted via a Mindray Auto Hematology Analyzer (Mindray Biomedical Electronics Co. Ltd., China), with quality control measures rigorously followed according to the manufacturer’s guidelines.
Mid‒upper arm circumference was measured via non-stretchable measuring tape according to the WHO recommendations to assess the participants’ nutritional status. Blood pressure was measured at the endpoint after iron folate supplementation in both arms, in a position where the brachial artery at the antecubital fossa is at the heart level.
Data were collected by the investigators assigned at each chosen hospital. To ensure the completeness, accuracy and consistency of data collection and for the investigators to have a common understanding of how to approach the participants, training was given before the start of data collection by the main investigator. On-site supervision was carried out weekly by the principal investigator. A pretest was performed at Abebech Gobena MCH Hospital, Addis Ababa, on 10% of the sample to check for the accuracy of the responses, language clarity, and appropriateness of the tools. Following the pretest, some adjustments were performed, including typing errors being fixed, data collectors being reoriented, and questionnaires being rearranged.
Data processing and analysis
The completeness and consistency of the data were checked. The data were then entered into EpiData version 4.6 software. After the data were edited and coded, they were exported to SPSS version 27 software for further analysis. Descriptive statistics were used to describe the profile of the study participants and to determine the hematological indices of the pregnant women enrolled in the study. Ninety-five percent CIs and corresponding P values < 0.05 were used to declare statistical significance. A logistic regression model was fitted to identify the associated variables.